[1]魏洪健,徐琦琳,张 瑛,等.一种基于 ZYNQ 的视频采集处理系统[J].计算机技术与发展,2022,32(11):24-29.[doi:10. 3969 / j. issn. 1673-629X. 2022. 11. 004]
　WEI Hong-jian,XU Qi-lin,ZHANG Ying,et al.A Video Acquisition and Processing System Based on ZYNQ[J].,2022,32(11):24-29.[doi:10. 3969 / j. issn. 1673-629X. 2022. 11. 004]
点击复制

一种基于 ZYNQ 的视频采集处理系统()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年11期
页码:
24-29
栏目:
嵌入式计算
出版日期:
2022-11-10

文章信息/Info

Title:
A Video Acquisition and Processing System Based on ZYNQ
文章编号:
1673-629X(2022)11-0024-06
作者:
魏洪健1 ; 徐琦琳1 ; 张 瑛1 ; 孙科学1; 2*
1. 南京邮电大学 电子与光学工程学院,江苏 南京 210023;
2. 射频集成与微组装技术国家地方联合工程实验室,江苏 南京 210023
Author(s):
WEI Hong-jian1 ; XU Qi-lin1 ; ZHANG Ying1 ; SUN Ke-xue1; 2*
1. School of Electronic and Optical Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210023,China;
2. Nation-Local Joint Project Engineering Lab of RF Integration & Micropackage,Nanjing 210023,China
关键词:
ZYNQ; ARM 处理器; 图像预处理; 实时视频处理; 现场可编程门阵列
Keywords:
ZYNQ; ARM processor; image preprocessing; real-time video processing; field programmable gate array
分类号:
TP274+. 2
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 11. 004
摘要:
针对当前基于 ARM 与 DSP 的图像处理框架存在的带宽低、速度慢的问题,提出了一种 ARM+FPGA 异构的 ZYNQ-7020 芯片作为主控的视频实时处理框架。 该系统由 500 万像素级别 CMOS 摄像头 OV5640 采集 1 280伊720 分辨率下 60 帧图像数据,处理后通过 HDMI 高清接口进行输出。 系统搭建了三种图像滤波处理,图像边缘检测及增强,形态学腐蚀与膨胀等图像预处理算法模块,均采用流水结构最大化处理速度以保证视频信息的实时性,封装为通用视频接口 IP 核,有较好的可移植性与扩展性。 同时系统集成了 RTL8211E 路由芯片,可通过 UDP / IP 协议对内部算法模块进行控制,切换实时输出的算法通道。 搭建 Vivado 与 Modelsim 联合仿真环境,对系统的实时性及算法有效性进行验证。 结果表明,系统可完成对图像的实时采集、处理、输出功能,可很好地抑制摄像头前端采集时产生的噪点,相比传统 ARM 端图像处理速度有较大的提升。
Abstract:
The current image processing framework based on ARM and DSP has the disadvantages of low bandwidth and slow speed. Toaddress this problem,a video real-time processing framework based on ARM+FPGA heterogeneous Zynq-7020 chip is proposed. The 5million pixel level CMOS camera OV5640 collects 60 frames of image data at a resolution of 1 280伊720,then outputs the video streamafter being processed through the HDMI high-definition interface. The system integrates lots of image preprocessing algorithm modulessuch as three kinds of image filtering processing, image edge detection, sharpen, morphological corrosion and dilation. All the aboveadopt a pipeline structure to maximize the processing speed to ensure the real - time performance of video information, packaged as ageneral video interface IP core to gain better performance on flexibility and scalability. The system also integrates the RTL8211Erouting chip,which can control the internal algorithm module through the UDP / IP protocol and switch the output algorithm channel. Toverify the real-time performance of the system and the effectiveness of the algorithm through setting up a joint simulation environment ofVivado and Modelsim,it is concluded that the system can complete the functions of real-time image acquisition,processing and output,and can well suppress the noise generated when the camera front acquisition, and the image processing speed is greatly improvedcompared with the traditional ARM end.

相似文献/References:

[1]邢艳芳,朱金付,周晓梅.基于Zynq 多核运行设计[J].计算机技术与发展,2018,28(03):60.[doi:10.3969/ j. issn.1673-629X.2018.03.012]
　XING Yan-fang,ZHU Jin-fu,ZHOU Xiao-mei.Design of Multi-core Processing Based on Zynq[J].,2018,28(11):60.[doi:10.3969/ j. issn.1673-629X.2018.03.012]
[2]许 杰,张子恒,王新宇,等.一种基于 Zynq 的 CNN 加速器设计与实现[J].计算机技术与发展,2021,31(11):108.[doi:10. 3969 / j. issn. 1673-629X. 2021. 11. 018]
　XU Jie,ZHANG Zi-heng,WANG Xin-yu,et al.Design and Implementation of CNN Accelerator Based on Zynq[J].,2021,31(11):108.[doi:10. 3969 / j. issn. 1673-629X. 2021. 11. 018]
[3]尹飞皓,尚建华.基于 ZYNQ 平台的图像处理算法 IP 验证系统研究[J].计算机技术与发展,2022,32(10):41.[doi:10. 3969 / j. issn. 1673-629X. 2022. 10. 007]
　YIN Fei-hao,SHANG Jian-hua.Research on IP Verification System of Image Processing Algorithm Based on ZYNQ Platform[J].,2022,32(11):41.[doi:10. 3969 / j. issn. 1673-629X. 2022. 10. 007]

更新日期/Last Update: 2022-11-10